Epitaxial growth of a chirped superlattice with constant dopings is achieved with minimal growth interruption time. This is done by doping only one of the two compositions during growth of its layer. For example, in the growth of a plurality of alternating layers of InP and GaInAs to form the superlattice, either the InP layers are doped with an n-type dopant, such as silicon, or the GaInAs layers are doped with a p-type dopant, such as beryllium. Alternatively, InP can equally be doped p-type (with beryllium) and GaInAs can be doped n-type (with silicon). In either case, the doping scheme described herein is easily done during molecular beam epitaxial growth by opening and closing the shutter of the dopant (silicon or beryllium) source cell at the appropriate times. To the electrical carriers, the doping superlattice scheme of the present invention presents a uniform doping without any need to change the doping cell temperature. This in turn allows improved controllability and minimal interface contamination.
|
1. A method of doping a region in a semiconductor device having a graded compositional alloy to provide a constant doping level, said method comprising forming said region sandwiched between two separate layers, one having a first composition and the other having a second composition, said graded compositional alloy comprising a plurality of alternating layers of said first composition and said second composition, wherein one of said compositions is doped with a dopant, wherein said doping of said one of said compositions is conducted for an amount of time which is essentially identical for each layer of said composition in said plurality of alternating layers.
13. A method of doping a region in a semiconductor device having a superlattice comprising an n-part and a p-part both having a constant doping level, said method comprising forming said region sandwiched between two separate layers, one having a first composition and the other having a second composition, said superlattice comprising a plurality of alternating layers of said first composition and said second composition, wherein only one of said compositions is doped with a dopant in said n-part and only one of said compositions is doped with a dopant in said p-part, wherein said doping of said one of said compositions in said n-part is conducted for an amount of time which is essentially identical for each layer of said composition in said n-part and said doping of said one of said compositions in said p-part is conducted for an amount of time which is essentially identical for each layer of said composition in said p-part.
19. A method of doping a region in a semiconductor device having a superlattice comprising an n-part and a p-part both having a constant doping level, said method comprising forming said region sandwiched between two separate layers, one having a first composition and the other having a second composition, said superlattice comprising a plurality of alternating layers of said first composition and said second composition, wherein only one of said compositions is doped with a dopant in said n-part and only one of said compositions is doped with a dopant in said p-part, wherein said doping of said one of said compositions in said n-part is conducted by opening a shutter of a dopant source for an amount of time which is essentially identical for each layer of said composition in said n-part and said doping of said one of said compositions in said p-part is conducted by opening a shutter of a dopant source for an amount of time which is essentially identical for each layer of said composition in said p-part.
2. The method of
3. The method of
4. The method of
7. The method of
8. The method of
9. The method of
10. The method of
11. The method of
12. The method of
14. The method of
15. The method of
16. The method of
20. The method of
21. The method of
22. The method of
|
|||||||||||||||||||||||||||
The present application is related to application Ser. No. 08,349,096, filed on even date herewith [PD-94179]. That application is directed to the use of parabolic compositional grading of the base-collector in a double heterojunction bipolar transistor. The present application is directed to the provision of an effective constant doping of the base-collector, whether the compositional profile is linear or parabolic or other profile.
1. Field of the Invention
The present invention relates generally to III-V semiconductor devices, and, more particularly, to compound semiconductors in a graded superlattice.
2. Description of Related Art
A graded compound semiconductor is usually implemented with molecular beam epitaxy as a gradient ("chirped") superlattice. For example, to grade from InP to GaInAs, one would typically have a short period (say 30 Å) superlattice where the thickness of InP component decreases from the full period to zero while the thickness of the GaInAs component increases correspondingly from zero to the full period when going from the InP to the GaInAs end. Doping such a chirped superlattice uniformly is difficult because the two components have in general different growth rates.
The growth rate, doping flux and bulk doping are related through the relationship:
bulk doping=doping flux/growth rate.
Since the growth rates of the components in the gradient (chirped) superlattice are usually different, the designer is forced to accept one of the following compromises:
(1) Maintain a constant doping flux and therefore different dopings in the different superlattice components due to the growth rate differences. However, this approach results in an effective doping gradient, discussed with respect to FIG. 2, below.
(2) Insert extensive growth interruptions (say 2 minutes) every time a different component is being grown. This allows time to change the doping cell temperature for a corrected doping flux, and is discussed with respect to FIG. 3, below. However, this approach unnecessarily exposes the growth surface to possible background contamination in the growth chamber.
No prior art is known to the inventors that is related to achieving a constant doping in a gradient (chirped) superlattice. However, the foregoing compromises are well-known to those skilled in the molecular beam epitaxy art.
Thus, a doping method which provides an essentially constant concentration throughout the chirped superlattice is required.
In accordance with the invention, epitaxial growth of a chirped superlattice with constant dopings is achieved with minimal growth interruption time. This is done by doping only one of the two compositions during growth of its layer. For example, in the growth of a plurality of alternating layers of InP and GaInAs to form the superlattice, either the InP or the GaInAs layers are doped with a dopant. The dopant is silicon for n-type doping and beryllium for p-type doping in the case of both InP and GaInAs. The doping is easily done during molecular beam epitaxial growth by opening and closing the shutter of the dopant (silicon or beryllium) source cell at the appropriate times.
To the electrical carriers, the doping superlattice scheme of the present invention presents a uniform doping without any need to change the doping cell temperature. This in turn allows improved controllability and minimal interface contamination.
FIG. 1 is a schematic illustration of the conduction bandedge of a chirped superlatice graded from InP to GaInAs;
FIG. 2 is a schematic illustration of the doping profile of the chirped superlattice of FIG. 1, resulting from employing a constant doping flux, and illustrating the effective doping concentration;
FIG. 3 is a schematic illustration of the chirped superlattice of FIG. 1, but with a growth interruption at every hetero-interface to accommodate the time required to change the temperature of the dopant cell in order to achieve a constant effective doping concentration profile;
FIG. 4a is a schematic illustration of the chirped superlattice of FIG. 1, and is used as an aid in understanding FIG. 4b
FIG. 4b illustrates the resultant effective constant doping achieved by the process of the invention;
FIG. 5 is a cross-sectional view of the device obtained by the process of the invention; and
FIG. 6, on coordinates of current (I, in microAmps) and voltage (V, in volts), are plots of output characteristics.
A compositional graded compound semiconductor is usually realized as a chirped superlattice in gas-source molecular beam epitaxy (GSMBE). In the following discussion, a description of a particular materials system is provided, namely, the grading from InP to GaInAs. However, the discussion is applicable to the doping of any graded compound where the growth rates of the two target materials are different. FIG. 1 shows a schematic illustration of the conduction bandedge of a chirped superlattice graded from InP to GaInAs. Curve 10 represents the composition that is grown, here, comprising a plurality of alternating layers of InP and GaInAs, with one period 12 represented by a pair of InP and GaInAs layers. It will be seen that at the left side of the Figure, the InP layer 10a is comparatively thicker than the GaInAs layer 10b. The first InP layer 10a is formed on an underlying InP layer (see FIG. 5). As successive periods 12 are grown, the ratio of layer thicknesses changes, so that at the final period, the GaInAs layer 10b is comparatively thicker than the InP layer 10a. The effective conduction band-edge, Ec, is represented by Curve 14, shown here as linear.
Typically, the growth rate of GaInAs is 1 μm/hr and that of InP is 0.532 μm/hr. If a constant doping flux (and therefore minimal growth interruption) is used, then a non-intentional doping grade is developed. FIG. 2 depicts a schematic illustration for the resulting doping. The level of doping in InP is represented at 15a, while the level of doping in GaInAs is represented at 15b. The doping concentration in InP is higher than in GaInAs because of the lower growth rate. Curve 16, which is the effective doping concentration, is seen to be graded.
Another approach is to change the doping cell temperature at every interface: The grower would insert growth interruption at every hetero-interface and re-adjust the doping cell temperature. FIG. 3 depicts the results of this growth; the arrows 18 indicate the locations of all the hetero-interfaces. The extensive growth interruptions cause background contamination to be incorporated at the interfaces unnecessarily.
In accordance with the invention, the doping is confined to only one of the two components in the superlattice, but the doping is increased correspondingly so that the weighted average of the doping is at the intended doping level. The doping cell temperature is not changed. See FIG. 4 for a schematic illustration where only the InP is being doped. Since only one component is being doped, there is no need to re-adjust the doping cell temperature.
The doping level of the doping superlattice required so that the weighted average is at the intended doping level is determined by the following equation:
Ds =(Ps /t)*d
where Ds is the superlattice doping, Ps is the superlattice period, t is the thickness of the doped layer in each period, and d is the intended doping level.
The present invention provides a novel doping scheme (in the form of a doping superlattice) such that the chirped superlattice appears to be uniformly doped to the carriers.
In one embodiment, as the InP thickness is changing from one period 12 to the next, as discussed above, the doping (silicon or beryllium) shutter is kept open the same amount of time for each InP layer. The result is that there is a fixed amount of donors (in the case of silicon) or acceptors (in the case of beryllium) here, electrons (in the case of silicon) or holes (in the case of beryllium in the InP layer. To the carriers, this appears to be a "constant" doping. The constant doping of each InP layer is shown in FIG. 4b as layers 20 of constant thickness; the resulting doping is effectively constant, as shown by line 22.
To test whether this concept works or not, the doping scheme of the invention was incorporated into an actual device that requires a constant doping in a graded layer for proper operation. It will be appreciated by those skilled in this art that direct measurement of the effective constant doping is very difficult. The device under test had a chirped superlattice grade between the InP region an the GaInAs region, similar to that illustrated in FIG. 1, but with the thicknesses of the individual layers designed to result in an effective parabolic compositional grade, and required a constant p and a constant n doping.
FIG. 5 provides a cross-sectional view of the device 24. For clarity, the patterning of the device 24 to expose the base, collector, an emitter contact layers and the fabrication of ohmic contacts to the contact layers is omitted. A semi-insulating substrate 26 comprises InP. A collector contact layer 28 is formed on the substrate and comprises n-doped GaInAs, 700 nm thick and doped t a concentration of 1×1019 cm-3. A subcollector layer 30 is formed on the collector contact layer 28 and comprises n-doped InP, 100 nm thick and doped to a concentration of 1×1019 cm-3. A collector layer 32 is formed on the subcollector layer 30 and also comprises n-doped InP, 750 nm thick and doped to a concentration of 2.4×1016 cm-3. A base-collector grading layer 34 is formed on the collector layer 32 and employs the doping superlattice of the present invention, starting with n-doped (silicon) InP (concentration at 1×1017 cm-3), 50 nm, and concluding with p-doped (beryllium) GaInAs (concentration at 1×1017 cm-3), 50 nm. Specifically, the n-doped part of the grade consists of a chirped superlattice having 17 periods, each period being about 3 nm thick, which results in a total thickness of about 50 nm for the n-doped region. The p-doped region is similarly constructed.
Next, a base spacer layer 36 is formed on the graded base-collector layer 34 and comprises p-doped GaInAs, 10 nm thick and doped to a concentration of 1×1017 cm-3. A base layer 38 is formed on the base spacer layer 36, also comprising p-doped GaInAs, 50 nm thick and doped to a concentration of 2×1019 cm-3. Another base spacer layer 40 is formed on the base layer 38, also comprising p-doped GaInAs, 10 nm thick and doped to a concentration of 1×1018 cm-3. An emitter-base superlattice grading layer 42 is formed on top of the base spacer layer 40, starting with p-doped GaInAs (concentration at 1×1018 cm-3), 9.9 nm, and concluding with n-doped AlInAs (concentration at 8×1017 cm-3), 19.8 nm. An emitter layer 44 is formed on the emitter-base grading layer 42, comprising n-doped AlInAs, 120 nm thick and doped to a concentration of 8×1017 cm-3. An emitter contact layer 46 is formed on the emitter layer 44, also comprising n-doped AlInAs, 70 nm thick and doped to a concentration of 1×1019 cm-3. A cap layer 48 is formed on the emitter contact layer 46 and comprises n-doped GaInAs, 100 nm thick and doped to a concentration of 1×10 19 cm-3.
But for the parabolic grade and the doping superlattice, the device 24 described above is a conventional double heterojunction bipolar transistor (DHBT). Thus, the thicknesses and dopant concentrations are to be considered as exemplary only.
The proper operation of the device 24 requires a well-controlled doping dipole; that is, the effective dopings in the graded regions must be constant. As can be seen from FIG. 6, the output characteristics has a very low knee voltage and minimal kink characteristics. Specifically, FIG. 6 shows the output collector current as a function of collector voltage, for different values of base current (in steps of 20 μA), in common-emitter configuration.
The process of the present invention is expected to have a significant impact on a number of microwave devices, including, but not limited to, (1) analog-to-digital (A/D) devices; (2) radar system very-high-frequency (VHF) switches operating above 200 MHz at>200 V, radar system wideband (7 to 11 GHz); (3) high power (>8 W), high gain (>9 dB), high efficiency (>50%) power amplifiers with heterojunction bipolar transistor (HBT) of breakdown voltage, open-circuited collector-base (BVcbo)>25 V maximum frequency of oscillation (Fmax)>90 GHz power cell 2 W @ 50% power-added efficiency (PAE); (4) V-band power devices; and (5) C- and Ku-band power devices.
Thus, there has been disclosed a method of obtaining an effective constant doping in a graded compositional alloy. It will be readily appreciated by those skilled in this art that changes or modifications of an obvious nature may be made, and all such changes and modifications are considered to fall within the scope of the invention as defined by the appended claims.
| Patent | Priority | Assignee | Title |
| 10056340, | May 31 2013 | HRL Laboratories, LLC | Flexible electronic circuit and method for manufacturing same |
| 6320212, | Sep 02 1999 | HRL Laboratories, LLC. | Superlattice fabrication for InAs/GaSb/AISb semiconductor structures |
| 6563145, | Apr 19 1999 | Skyworks Solutions, Inc | Methods and apparatus for a composite collector double heterojunction bipolar transistor |
| 6653213, | Dec 21 2000 | NORTEL NETWORKS UK LIMITED | Structure and method for doping of III-V compounds |
| 6670653, | Jul 30 1999 | HRL Laboratories, LLC | InP collector InGaAsSb base DHBT device and method of forming same |
| 6727153, | Sep 02 1999 | HRL Laboratories, LLC | Superlattice fabrication for InAs/GaSb/AlSb semiconductor structures |
| 7576409, | Aug 20 2004 | HRL Laboratories, LLC | Group III-V compound semiconductor based heterojuncton bipolar transistors with various collector profiles on a common wafer |
| 8216910, | Aug 20 2004 | HRL Laboratories, LLC | Group III-V compound semiconductor based heterojunction bipolar transistors with various collector profiles on a common wafer |
| 8697532, | Feb 07 2005 | HRL Laboratories, LLC | InP based heterojunction bipolar transistors with emitter-up and emitter-down profiles on a common wafer |
| 9530708, | May 31 2013 | HRL Laboratories, LLC | Flexible electronic circuit and method for manufacturing same |
| Patent | Priority | Assignee | Title |
| 4794611, | Dec 18 1984 | Canon Kabushiki Kaisha | Semiconductor laser having superlattice structure |
| 5091756, | Feb 16 1987 | Tokyo Institute of Technology | Superlattice structure |
| 5170407, | Oct 11 1991 | CHASE MANHATTAN BANK, AS ADMINISTRATIVE AGENT, THE | Elimination of heterojunction band discontinuities |
| 5185647, | Dec 12 1988 | The United States of America as represented by the Administrator of the | Long wavelength infrared detector |
| 5198682, | Nov 12 1991 | Hughes Aircraft Company | Multiple quantum well superlattice infrared detector with graded conductive band |
| 5289486, | Feb 28 1991 | Omron Corporation | Semiconductor luminous element and superlattice structure |
| 5296720, | Nov 17 1991 | HE HOLDINGS, INC , A DELAWARE CORP ; Raytheon Company | Apparatus and method for discriminating against undesired radiation in a multiple quantum well long wavelength infrared detector |
| 5324959, | Dec 08 1989 | Hitachi, Ltd.; Hitachi Device Engineering Co., Ltd. | Semiconductor optical device having a heterointerface therein |
| 5365077, | Jan 22 1993 | Hughes Electronics Corporation | Gain-stable NPN heterojunction bipolar transistor |
| 5386156, | Aug 27 1993 | Lattice Semiconductor Corporation | Programmable function unit with programmable fast ripple logic |
| 5420059, | Dec 22 1992 | International Business Machines Corporation | Method of making a high performance MESFET with multiple quantum wells |
| 5528051, | Jan 22 1992 | Picogiga Societe Anonyme | Quantum well infrared detector |
| JP2246381, | |||
| JP4282825, | |||
| JP53160, | |||
| JP6244115, |
| Executed on | Assignor | Assignee | Conveyance | Frame | Reel | Doc |
| Nov 30 1994 | LIU, TAKYIU | Hughes Aircraft Company | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 007304 | /0084 | |
| Nov 30 1994 | NGUYEN, CHANH | Hughes Aircraft Company | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 007304 | /0084 | |
| Dec 01 1994 | Hughes Electronics Corporation | (assignment on the face of the patent) | / | |||
| Dec 15 1997 | HE HOLDINGS INC , D B A HUGHES ELECTRONICS, FORMALLY KNOWN AS HUGHES AIRCRAFT COMPANY | Hughes Electronics Corporation | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 008935 | /0912 |
| Date | Maintenance Fee Events |
| Jun 25 2001 | ASPN: Payor Number Assigned. |
| Jun 25 2001 | RMPN: Payer Number De-assigned. |
| Nov 16 2001 | M183: Payment of Maintenance Fee, 4th Year, Large Entity. |
| Dec 11 2001 | REM: Maintenance Fee Reminder Mailed. |
| Nov 21 2005 | M1552: Payment of Maintenance Fee, 8th Year, Large Entity. |
| Nov 19 2009 | M1553: Payment of Maintenance Fee, 12th Year, Large Entity. |
| Date | Maintenance Schedule |
| May 19 2001 | 4 years fee payment window open |
| Nov 19 2001 | 6 months grace period start (w surcharge) |
| May 19 2002 | patent expiry (for year 4) |
| May 19 2004 | 2 years to revive unintentionally abandoned end. (for year 4) |
| May 19 2005 | 8 years fee payment window open |
| Nov 19 2005 | 6 months grace period start (w surcharge) |
| May 19 2006 | patent expiry (for year 8) |
| May 19 2008 | 2 years to revive unintentionally abandoned end. (for year 8) |
| May 19 2009 | 12 years fee payment window open |
| Nov 19 2009 | 6 months grace period start (w surcharge) |
| May 19 2010 | patent expiry (for year 12) |
| May 19 2012 | 2 years to revive unintentionally abandoned end. (for year 12) |